Computerized systems have made a significant contribution towards the advancement of modern society and are utilized in a number of applications to achieve advantageous results. The computerization of filing systems (i.e., systems for the storage and organization of documents or objects as electronic data) is one such example of an application that has benefited from the utilization of computerized systems.
Computerized filing systems are now available that enable an electronic object to be shared and updated among multiple remote systems. Popular applications of synchronized computerized filing systems include applications which monitor shared files for updates to any of the files being shared. Typically, when an object—such as a file—has been updated at any terminal with access, the update and/or the updated file is then perpetuated across the remaining distributed replicas of the file by a synchronizing controller application.
There are multiple file systems in existence on various platforms (e.g., FAT, NTFS, HFS, etc.) and operating systems. File synchronization software needs to watch for changes to files and folders (and other artifacts) on these file systems that are being synchronized. However, not only do different operating systems have different methods of interacting with the underlying file system, the different file systems also have different capabilities, metadata and application programming interface (“API”) support for detecting changes. Existing synchronization software typically designate only one file system to support and do not extend synchronization functionality to other file systems. Synchronization software that support multiple platforms tend to use the least common denominator approach—providing only compatible, often basic capabilities—and therefore cannot make use of advanced capabilities on some file systems.
Additional problems may arise when applications relate distinct files together while operating on them. For example, a database application may store data in both a data file and an index file-one of these files is not enough to recreate or interpret all the data. During synchronization, the versions of both files must match for the application to work. This inevitably adds to the complexity of the synchronization operation.
The file manipulation processes for some application may include creating temporary hidden files and performing rename and delete operations during subsequent updates to obtain the final operations. In such a process, the watcher must use rules to ignore the intermediate renames and deletes, etc. Currently implemented watchers have either a few of these rules statically coded or none at all, and as a result suffer from a lack of efficiency and/or flexibility to deal with new idiosyncrasies in file manipulation techniques.